Output details
15 - General Engineering
University of Cambridge
Structured Log Linear Models for Noise Robust Speech Recognition
This paper formed the basis for a number of additional papers including a paper short-listed for the best student paper award at the IEEE conference on Acoustics, Speech and Signal Processing 2011, and an article in IEEE Signal Processing Magazine "Structured Discriminative Models for Speech Recognition", November 2012. The work, forms in part the basis of an EPSRC funded project "Generative Kernels and Score-Spaces for Classification of Speech", EP/I006583/1. This project aims to combine discriminative and generative models together to improve the performance of speech recognition systems in highly challenging environments, such as in the presence of background noise.